Web Usage Mining: A Survey on Pattern Extraction from Web Logs

Web Usage Mining: A Survey on Pattern Extraction from Web Logs  

Web Usage Mining: A Survey on Pattern Extraction from Web Logs 1

S. K. Pani, , 2L. Panigrahy, 2V.H.Sankar, 3Bikram Keshari Ratha, 2A.K.Mandal, 2S.K.Padhi 1 P.G. Department Of Computer Science, RCMA; Bhubaneswar, Orissa, India 2 Department of Computer Science and Engineering; Konark Institute of Science and Technology; Bhubaneswar, Orissa, India 3 P.G. Department Of Computer Science, Utkal University,Bhubaneswar, Orissa, India E-mail: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]

Abstract— As the size of web increases along with number of users, it is very much essential for the website owners to better understand their customers so that they can provide better service, and also enhance the quality of the website. To achieve this they depend on the web access log files. The web access log files can be mined to extract interesting pattern so that the user behaviour can be understood. This paper presents an overview of web usage mining and also provides a survey of the pattern extraction algorithms used for web usage mining.

To mine the interesting data from this huge pool, data mining techniques can be applied. But the web data is unstructured or semi structured. So we can not apply the data mining techniques directly. Rather another discipline is evolved called web mining which can be applied to web data. Web mining is used to discover interest patterns which can be applied to

Keywords— web mining, pattern extraction, usage mining, preprocessing

many real world problems like improving web sites, better I. INTRODUCTION In this world of

Information

understanding the visitor’s behavior, product recommendation

Technology, accessing

etc.

information is the most frequent task. Every day we have to go through several kind of information that we need and what

Web mining is the use of data mining techniques to

we do? Just browse the web and the desired information is

automatically discover and extract information from Web

with us on a single click. Today, internet is playing such a

documents/services

vital role in our everyday life that it is very difficult to survive

categorized into 3 types. 1. Content Mining (Examines the

without it. The World Wide Web (WWW) has influenced a lot

content of web pages as well as results of web Searching) 2.

to both users (visitors) as well as the web site owners. The

Structure Mining (Exploiting Hyperlink Structure) 3. Usage

web site owners are able to reach to all the targeted audience

Mining (analyzing user web navigation)

(Etzioni,1996).

Web

mining

is

nationally and internationally. They are open to their customer Web usage mining is a process of picking up information from

24X7. On the other side visitors are also availing those

user how to use web sites. Web content mining is a process of

facilities.

picking up information from texts, images and other contents. In the last fifteen years, the growth in number of web sites and

Web structure mining is a process of picking up information

visitors to those web sites has increased exponentially. The

from linkages of web pages.

number of users by June 30 2010 was 1,966,514,816[18] Web Mining 

which is 28.7% of the world’s population. The number of active web sites is 125,522,259 [19] as on 13-Dec-2010. Due to this growth a huge quantity of web data has been generated.

Content Mining

Structure Mining 

Usage Mining

Figure 1: Web mining classification

  15 International Journal of Instrumentation, Control & Automation (IJICA), Volume 1, Issue 1, 2011

   

Web Usage Mining: A Survey on Pattern Extraction from Web Logs  

3.

Data preprocessing – Cleaning and structuring data to prepare for pattern extraction

4.

Pattern extraction – Extracting interesting patterns

5.

Pattern analysis and visualization – Analyze the extracted pattern

6.

Pattern applications – Apply the pattern in real world problems

TABLE 1  THE RELATIONSHIP AMONG THE DIFFERENT AREAS OF WEB MINING

Type

Structure

Form

Object

Collection

Usage

Accessing

Click

Behavior

Logs

Content

Pages

Text

Index

Pages

Structure

Map

Hyperlinks

Map

Hyperlinks

II. RELATED WORK ON PATTERN EXTRACTION The existing research works done by different authors can be

These 3 approaches attempts to extract knowledge from Web generate some useful result from that knowledge and apply

categorized into a) Association Rule Mining (ARM), b)

the result to certain real world problems. Web Usage Mining

Clustering, c) Classification

is the process of applying data mining techniques to the

TABLE 2 

discovery of usage patterns from data extracted from Web

THE RESEARCH WORK DONE ON ARM

Log files.

Algorithm Used

Authors

Year 1998

Maximal

Ming-Syan Chen, Jong Soo

Web usage mining is one of the prominent research area due

forward references

Park, Philip S. Yu

to these following reasons. a) One can keep track of

Markov Chains

Jianhan Zhu, Jun Hong, and

previously accessed pages of a user. These pages can be used

2002

John G. Hughes

to identify the typical behavior of the user and to make

Improved AprioriAll

WANG Tong, HE Pi-lian

2005

Fpgrowth and

Hengshan Wang,

2006

Prefixspan

Cheng Yang,

prediction about desired pages. Thus personalization for a user can be achieved through web usage mining. b) Frequent access behavior for the users can be used to identify needed

Hua Zeng

links to improve the overall performance of future accesses. Prefetching and caching policies can be made on the basis of

Custom Built

Sandeep Singh Rawat,

APRIORI Algorithm

Lakshmi Rajamani

2010

frequently accessed pages to improve latency time. c) Common access behaviors of the users can be used to improve

Ming-Syan et al., [6] proposed a new data mining algorithm

the actual design of web pages and for making other

that involves mining path traversal patterns in a distributed

modifications to a Web site. d) Usage patterns can be used for

information-providing environment where documents or

business intelligence in order to improve sales and

objects are linked together to facilitate interactive access.

advertisement by providing product recommendations.

Their solution procedure consists of two steps. First, derive an algorithm to convert the original sequence of log data into a

Five major steps followed in web usage mining are

set of maximal forward references. Second, derive another 1.

Data collection – Web log files, which keeps track of visits of all the visitors

2.

Data Integration – Integrate multiple log files into a single file

algorithms to determine the frequent traversal patterns i.e., large reference sequences from the maximal forward references obtained. Jianhan Zhu et al., [7] applied the Markov chains to model user navigational behavior. They

  16 International Journal of Instrumentation, Control & Automation (IJICA), Volume 1, Issue 1, 2011

   

Web Usage Mining: A Survey on Pattern Extraction from Web Logs  

proposed a method for constructing a Markov model of a web site based on past visitor behavior. Then the Markov model is

Paola Britos et al., [10] described the capacity of use of Self

used to make link predictions that assist new users to navigate

Organized Maps, kind of artificial neural network, in the

the Web site. WANG Tong et al., [8] offers an improved

process of Web Usage Mining to detect user’s patterns. The

algorithm based on the original AprioriAll algorithm. The new

process detail the transformations necessaries to modify the

algorithm adds the property of the UserID during the every

data storage in the Web Servers Log files to an input of Self

step of producing the candidate set and every step of scanning

Organized Maps. Mehrdad Jalali et al., [11] presented an

the database by which to decide whether an item in the

approach which is based on the graph partitioning for

candidate set should be put into the large set which will be

modeling user navigation patterns. In order to mining user

used to produce next candidate set. Hengshan Wang et al., [9]

navigation patterns, they establish an undirected graph based

introduced two prevalent data mining algorithms - FPgrowth

on connectivity between each pair of the web pages and also

and PrefixSpan into WUM. Maximum Forward Path (MFP) is

proposed novel formula for assigning weights to edges of the

also used in the web usage mining model during sequential

graph. Kobra Etminani et al., [12]

pattern mining along with PrefixSpan so as to reduce the

clustering algorithm to pre-processed logs to extract frequent

interference of “false visit” caused by browser cache and raise

patterns for pattern discovery and then it is displayed in an

the of mining frequent traversal paths. Sandeep Singh Rawat

interpretable format. N. Sujatha et al., [15] have proposed a

et al., [13] proposed a custom-built apriori algorithm which is

new framework to improve the web sessions’ cluster quality

based on the old Apriori algorithm, to find the effective

from k-means clustering using Genetic Algorithm (GA).

pattern analysis

Initially a modified k-means algorithm is used to cluster the

applied ant-based

TABLE 3 

user sessions. The refined initial starting condition allows the

THE RESEARCH WORK DONE ON CLUSTERING

iterative algorithm to converge to a “better” local minimum. And in the second step, they have proposed a GA based

Algorithm Used

Authors

Year

refinement algorithm to improve the cluster quality. TABLE 4 

Paola Britos, Damián Self Organized Maps

Martinelli, Hernán Merlino,

THE RESEARCH WORK DONE ON CLASSIFICATION

2007

Ramón García-Martínez Mehrdad Jalali, Norwati Graph Partitioning

Mustapha, Ali Mamat, Md.

Algorithm Used

Authors

Year

Naive Bayesian

Mahdi Khosravi, Mohammad J.

2010

Tarokh

2008

Nasir B Sulaiman Kobra Etminani,

Mahdi Khosravi et al., [14] proposed a novel approach for

Mohammad-R. Ant-based

Akbarzadeh-T, Noorali

dynamic mining of users’ interest navigation patterns, using

2009

naïve Bayesian method

Raeeji Yanehsari K-mean with Genetic Algorithm

III. BASIC CONCEPTS N. Sujatha, K. Iyakutty

2010

A. Data Mining

  17 International Journal of Instrumentation, Control & Automation (IJICA), Volume 1, Issue 1, 2011

   

Web Usage Mining: A Survey on Pattern Extraction from Web Logs  

There are two classes [4] of data mining namely i) to

relationships between occurrences of sequential events, to find

summarize or characterize general properties of data in

if there exists any specific order of the occurrences. We can

repository which is called Descriptive and

ii) to perform

find the sequential patterns of specific individual items; also

inference on current data, to make predictions based on the

we can find the sequential patterns cross different items.

historical data which is called Prescriptive. There are various

Sequential pattern mining is widely used in analyzing of DNA

data mining techniques available which also can be applied to

sequence. An example of sequential patterns is that every time

web data mining. Few techniques are listed below.

Microsoft stock drops 5%, IBM stock will also drops at least

1) Association Rules Mining: When the book Data Mining

4% within three days.

Concepts and Techniques is bought, 40% of the time the book

3) Classification: Classification is to build (automatically) a

Database System is bought together, and 25% of the time the

model that can classify a class of objects so as to predict the

book Data Warehouse is bought together. Those rules

classification or missing attribute value of future objects

discovered from the transaction database of the book store can

(whose class may not be known). It is a two-step process. In

be used to rearrange the way of how to place those related

the first process, based on the collection of training data set, a

books, which can further make those rules more strong

model is constructed to describe the characteristics of a set of

2) Sequential Pattern Mining: Association rule mining does

data classes or concepts. Since data classes or concepts are

not take the time stamp into account, the rule can be Buy

predefined, this step is also known as supervised learning (i.e.,

A=>Buy B. If we take time stamp into account then we can

which class the training sample belongs to is provided). In the

get more accurate and useful rules such as: Buy A implies

second step, the model is used to predict the classes of future

Buy B within a week, or usually people Buy A every week.

objects or data. A decision tree for the class of buy laptop,

As we can see with the second kind of rules, business

indicate whether or not a customer is likely to purchase a

organizations can make more accurate and useful prediction

laptop. Each internal node represents a decision based on the

and consequently make more sound decisions. A database

value of corresponding attribute, also each leaf node

consists of sequences of values or events that change with

represents a class (the value of buy laptop=Yes or No). After

time, is called a time-series database [Han and Kamber 2000],

this model of buy laptop has been built, we can predict the

a time-series database records the valid time of each dataset.

likelihood of buying laptop based on a new customer's

For example, in a time-series database that records the sales

attributes such as age, degree and profession. That information

transaction of a supermarket, each transaction includes an

can be used to target customers of certain products or services,

extra attribute indicate when the transaction happened. Time-

especially widely used in insurance and banking.

series database is widely used to store historical data in a

4) Clustering: Classification can be taken as supervised

diversity of areas such as, financial data, medical data,

learning process, clustering is another mining technique

scientific data and so on. Different mining techniques have

similar to classification. However clustering is a unsupervised

been designed for mining time-series data, basically there are

learning process. Clustering is the process of grouping a set of

four kinds of patterns we can get from various types of time-

physical or abstract objects into classes of similar objects, so

series data:1) Trend analysis, 2) Similarity search, 3)

that objects within the same cluster must be similar to some

Sequential patterns and 4) Periodical patterns. Sequential

extent, also they should be dissimilar to those objects in other

patterns: sequential pattern mining is trying to find the

clusters. In classification which record belongs which class is

  18 International Journal of Instrumentation, Control & Automation (IJICA), Volume 1, Issue 1, 2011

   

Web Usage Mining: A Survey on Pattern Extraction from Web Logs  

predefined, while in clustering there is no predefined classes.

with the # character contain directives. The following

In clustering, objects are grouped together based on their

directives are defined in the W3C Extended format [16, 17]:

similarities. Similarities between objects are defined by The following is an example of a record in the extended log

similarity functions, usually similarities are quantitatively

format that was produced by the Microsoft Internet

specified as distance or other measures by corresponding

Information Server (IIS):

domain experts. For example, based on the expense, deposit and draw patterns of the customers, a bank can clustering the

#Software: Microsoft Internet Information Server 4.0 #Version: 1.0 #Date: 1998-11-19 22:48:39 #Fields: date time c-ip cs-username s-ip cs-method cs-uri-stem cs-uri-query scstatus sc-bytes cs-bytes time-taken csversion cs(User-Agent) cs(Cookie) cs(Referrer)

market into different groups of people. For different groups of market, the bank can provide different kinds of loans for houses or cars with different budget plans. In this case the bank can provide a better service, and also make sure that all the loans can be reclaimed. B. Log Files In order to manage a web server effectively, it is necessary to

1998-11-19 22:48:39 206.175.82.5 208.201.133.173 GET /global/images/navlineboards.gif - 200 540 324 157 HTTP/1.0 Mozilla/4.0+(compatible;+MSIE+4.01;+Windo ws+95) USERID=CustomerA;+IMPID=01234 http://yourturn.rollingstone.com/webx?98@ @webx1.html

get feedback about the activity and performance of the server as well as any problems that may be occurring. Web server creates and maintains log files for this purpose [3]. A Web log is a file to which the Web server writes information each time a user requests a resource from that particular site.

Description of headers c Client s Server r Remote cs Client to Server. sc Server to Client. sr Server to Remote Server, this prefix is used by proxies. rs Remote Server to Server, this prefix is used by proxies. x Application specific identifier.

1) Log Formats: W3C maintains a standard format for web server log files, but other proprietary formats exist. For example IIS provides six different log file formats which are used to track and analyze information about IIS-based sites and services such as 1. W3C Extended Log File Format, 2. W3C Centralized Logging, 3. NCSA Common Log File Format, 4. IIS Log File Format, 5. ODBC Logging, 6. Centralized Binary Logging. In addition to the six available

Apache web server maintains Common Log Format and Combined Log Format[3].

formats, custom log file format can also be configured. A log file in the W3C extended format contains a sequence of

Common Log Format

lines containing ASCII characters. Each line may contain either a directive or an entry. Entries consist of a sequence of

LogFormat "%h %l %u %t \"%r\" %>s %b" common 122.163.111.210 - - [22/Oct/2010:04:15:03 -0400] "GET /imagesnew/misc_arrow_animated.gif HTTP/1.1" 404 494

fields relating to a single HTTP transaction. Fields are separated by white space. If a field is unused in a particular entry dash "-" marks the omitted field. Directives record information about the logging process itself. Lines beginning

Combined Log Format

  19 International Journal of Instrumentation, Control & Automation (IJICA), Volume 1, Issue 1, 2011

   

Web Usage Mining: A Survey on Pattern Extraction from Web Logs  

Another commonly used format string is called the Combined Log Format. It can be used as follows.

we make our site meeting user needs? Can we increase user

LogFormat "%h %l %u %t \"%r\" %>s %b

web content personalized to them? Answer to these questions

\"%{Referer}i\" \"%{User-agent}i\""

may come from the analysis of the data from log files stored

combined

in web servers. Web usage mining has then become a

This format is exactly the same as the Common Log Format,

necessary task in order to provide web administrators with

with the addition of two more fields. Each of the additional

meaningful information about users and usage patterns for

fields uses the percent-directive %{header}i, where header

improving

can be any HTTP request header. The access log under this

performance. Successful websites may be those that are

format will look like:

customized to meet user preferences both in the presentation

satisfaction? Can we target specific groups of users and make

quality

of

web

information

and

service

of information and in relevance of the content that best fits the

127.0.0.1 - frank [10/Oct/2000:13:55:36 0700] "GET /apache_pb.gif HTTP/1.0" 200 2326 "http://www.example.com/start.html" "Mozilla/4.08 [en] (Win98; I ;Nav)"

user. IV. STEPS IN WEB USAGE MINING

The additional fields are: A. Data Preparation The information contained in a raw Web server log does not

"http://www.silicon.ac.in/sitsbp/index.ht ml"

reliably represent a user session file [1]. The Web usage data

(\"%{Referer}i\") The "Referer" gives the site that the client reports having been referred from.

preparation phase is used to restore users' activities in the Web

"Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; GTB0.0; SV1; .NET CLR 2.0.50727; .NET CLR 3.0.04506.30; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; RediffIE8)"

at a minimum achieve the following four major tasks: i)

server log in a reliable and consistent way. This phase should

removing undesirable entries, ii) distinguishing among users, iii) building sessions, and iv) restoring the contents of a session.

(\"%{User-agent}i\")

1) Removing Undesirable Entries: Web logs contain user

The User-Agent HTTP request header. This is the identifying information that the client browser reports about itself.

activity information, of which some is not closely relevant to

 

usage mining and can be removed without noticeably

More recent entries are appended to the end of the file. These

affecting the mining, for example: a) All log image entries.

data can be stored into a single file, or separated into distinct

The HTTP protocol requires a separate connection for every

logs, such as an access log, error log, or referrer log.

file that is requested from the Web server. Images are

Web usage mining research focuses on finding patterns of

automatically downloaded based on the HTML page

navigational behavior from users visiting website. These

requested and the downloads are recorded in the logs. In the

patterns of navigational behavior can be valuable when

future, images may provide valuable usage information, but

searching answers to questions like: How efficient is our

the research on image understanding is still in the early stages.

website in delivering information? How the users perceive the

Thus, log image entries do not help the usage mining and can

structure of the website? Can we predict user’s next visit? Can

be removed. b) Robot assesses. A robot, also known as spider

  20 International Journal of Instrumentation, Control & Automation (IJICA), Volume 1, Issue 1, 2011

   

Web Usage Mining: A Survey on Pattern Extraction from Web Logs  

or crawler, is a program that automatically fetches Web pages.

7.50

Robots are used to feed pages to search engines or other software. Robot-access patterns can be identified from useragent so many of the robot accesses can be detected and removed from the logs.

:07:11 -0400

dept.php

c.in/departments.php

113.19.1 7.50

24/Oct/2010:03 :07:33 -0400

/cs-dept.php

http://www.silicon.a c.in/departments.php

113.19.1 7.50

24/Oct/2010:03 :08:21 -0400

/csfacdirectory. php

http://www.silicon.a c.in/cs-dept.php

113.19.1 7.50

24/Oct/2010:03 :08:44 -0400

/departments .php

http://www.silicon.a c.in/csfacdirectory.php

113.19.1 7.50

24/Oct/2010:03 :08:51 -0400

/bshdept.php

http://www.silicon.a c.in/departments.php

113.19.1 7.50

24/Oct/2010:03 :08:55 -0400

/bshfacdirectory. php

http://www.silicon.a c.in/bsh-dept.php

113.19.1 7.50

24/Oct/2010:03 :09:18 -0400

/departments .php

http://www.silicon.a c.in/bshfacdirectory.php

113.19.1 7.50

24/Oct/2010:03 :09:21 -0400

/it-dept.php

http://www.silicon.a c.in/departments.php

113.19.1 7.50

24/Oct/2010:03 :09:24 -0400

/itfacdirectory. php

http://www.silicon.a c.in/it-dept.php

2) Distinguishing among Users: A user is defined as a single individual that accesses files from web servers through a browser. A web log sequentially records users’ activities according to the time each occurred. In order to study the actual user behavior, users in the log must be distinguished. Figure 1 is a sample Web site where nodes are pages, edges are hyperlinks, and node “/index.php” is the entry page of this site. The edges are bi-directional because users can  easily use the back button on the browser to return to the previous page.

/index.php 

/bsh‐dept.php 

/bsh‐facdirectory.php

/it‐dept.php 

/it‐facdirectory.php

/cs‐dept.php 

/cs‐facdirectory.php

/departments.php 

Figure 2: Sample website structure

3) Building Sessions (sessionization / episode identification): For logs that span long periods of time, it is very likely that

The access data from an IP address (113.19.17.50) recorded

individual users will visit the Web site more than once or their

on the log are given in Table 5. The paths are found by

browsing

heuristics are index.php -- departments.php – bsh-dept.php –

may

be

interrupted.

The

goal

of

session

identification is to divide the page accesses of each user into

departments.php – cs-dept.php – cs-facdirectory.php –

individual sessions. A time threshold is usually used to

departments.php – bsh-dept.php – bsh-facdirectory.php –

identify sessions.

departments.php – it-dept.php -- it-facdirectory.php. 4) Restoring the Contents of a Session: This task determines TABLE 5 

if there are important accesses that are not recorded in the

SAMPLE LOG clientIP

reqDateTime

URL

referer

113.19.1 7.50

24/Oct/2010:02 :52:23 -0400

/index.php

-

113.19.1 7.50

24/Oct/2010:02 :52:29 -0400

/departments .php

http://www.silicon.a c.in/index.php

113.19.1

24/Oct/2010:03

/bsh-

http://www.silicon.a

access logs. For example, Web caching or using the back button of a browser will cause information discontinuance in logs. B. Pattern Discovery/ Pattern Extraction

  21 International Journal of Instrumentation, Control & Automation (IJICA), Volume 1, Issue 1, 2011

   

Web Usage Mining: A Survey on Pattern Extraction from Web Logs  

It deals with extracting interesting patterns from the pre

records two entries in access log. It is difficult to identify an

processed web logs. This is the key component of web usage

inner page. D) Some web pages take query string as argument

mining. This can be subdivided into

to the URL. E.g. dept.php?dept=CSE, dept.php?dept=IT like this. In this case the same page i.e. dept.php is accessed but

Statistical Analysis:

with different arguments. It is difficult to count the page

TABLE 6 

access of the web page without the argument.

STATISTICAL ANALYSIS CATEGORY

Category General statistics

Access statistics Periodical statistics

In web usage mining the pattern extraction algorithms are

Description 1) Total number of hits 2) Total number of visitors 3) Different errors 4) Successful visits 5) Incomplete visits 6) Error reports Request Hit and Miss count based on 1) IP address 2) URL Access of web pages according to period of time e.g. daily, monthly, yearly.

applied on the log data after they are processed. So preprocessing is very much important and must be carried out with proper care. While preprocessing the web access log the above points should be taken into consideration so that it will produce a good set of access logs for pattern extraction. VI. CONCLUSION There is a growing trend among companies, organizations and individuals alike to gather information through web mining to

Clustering: clustering can be done according to i) Web pages,

utilize that information in their best interest. But it is a

ii) Web page sequences, iii) client IP etc.

challenging task for them to fulfill the user needs and keep

Classification: classify users according to their navigational

their attention in their website. Web usage mining has

behavior.

valuable uses to the marketing of businesses and a direct

Association Rules: for example, thirty percent of department

impact to the success of their promotional strategies and

page viewers will enter the cs-dept pages.

internet traffic. This information is gathered on a daily basis and continues to be analyzed consistently. Analysis of this

V. CHALLENGES The web usage mining algorithms are applied on the

pertinent information will help companies to develop

preprocessed web log data. The log files are collected from

promotions that are more effective, internet accessibility,

web server. But there are certain reasons due to which the

inter-company communication and structure, and productive

actual logs are not collected. A) Due to the cache present on

marketing skills through web usage mining. If we will be able

client browser, most of the request, if it is present in the cache

to propose an efficient algorithm for the pattern extraction

are not sent to web server. B) Most of the time user does not

than it will help in the business of the website owners to

visit the home page of a website. They directly navigate to a

understand their customer’s behavior properly so that they can

particular page, by getting the URL from search engines. So it

fulfill their requirements.

reduces the hit count of index page. C) Generally in web pages designed by server side scripting like PHP, JSP or VI. REFERENCES

ASP.NET they use inner page. That is, one page consisting of

[1] Chen Hu, Xuli Zong, Chung-wei Lee and Jyh-haw Yeh, “World Wide Web Usage Mining Systems and Technologies”, Journal of SYSTEMICS, CYBERNETICS AND INFORMATICS Vol. 1, No. 4, Pages53-59, 2003.

more than one page. In that case the request for main page

  22 International Journal of Instrumentation, Control & Automation (IJICA), Volume 1, Issue 1, 2011

   

Web Usage Mining: A Survey on Pattern Extraction from Web Logs   [2] FlorentMasseglia, Pascal Poncelet, Rosine Cicchetti, “An efficient algorithm for Web usage mining”, Networking and Information Systems Journal. Volume X, 2000 [3] R. Pamnani, P. Chawan “Web Usage Mining: A Research Area in Web Mining” [4] Qiankun Zhao, Sourav S. Bhowmick, “Sequential Pattern Mining: A Survey”, Technical Report, CAIS, Nanyang Technological University, Singapore, No. 2003118 , 2003. [5] S. Rawat, L. Rajamani, “Discovering Potential User Browsing Behaviors Using Custom-Built APRIORI Algorithm”, International journal of computer science & information Technology (IJCSIT) Vol.2, No.4, August 2010 [6] Ming-Syan Chen, Jong Soo Park, Philip S. Yu, “Efficient Data Mining for Path Traversal Patterns”, IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 10, NO. 2, MARCH/APRIL 1998. [7] Jianhan Zhu, Jun Hong, John G. Hughes, “Using Markov Chains for Link Prediction in Adaptive Web Sites”, Soft-Ware 2002, LNCS 2311, pp. 60–73, 2002 [8] WANG Tong, HE Pi-lian, “Web Log Mining by an Improved AprioriAll Algorithm”, World Academy of Science, Engineering and Technology 4 2005 [9] Hengshan Wang, Cheng Yang, Hua Zeng, “ Design and Implementation of a Web Usage Mining Model Based On Fpgrowth and Prefixspan”, Communications of the IIMA 2006 Volume 6 Issue 2 [10] Paola Britos, Damián Martinelli, Hernán Merlino, Ramón GarcíaMartínez, “Web Usage Mining Using Self Organized Maps”, International Journal of Computer Science and Network Security, VOL.7 No.6, June 2007 [11] Mehrdad Jalali, Norwati Mustapha, Ali Mamat, Md. Nasir B Sulaiman, “WEB USER NAVIGATION PATTERN MINING APPROACH BASED ON GRAPH PARTITIONING ALGORITHM”, Journal of Theoretical and Applied Information Technology [12] Kobra Etminani, Mohammad-R. Akbarzadeh-T., Noorali Raeeji Yanehsari, “Web Usage Mining: users' navigational patterns extraction from web logs using Ant-based Clustering Method”, IFSA-EUSFLAT 2009 [13] Sandeep Singh Rawat, Lakshmi Rajamani, “DISCOVERING POTENTIAL USER BROWSING BEHAVIORS USING CUSTOM-BUILT APRIORI ALGORITHM”, International journal of computer science & information Technology (IJCSIT) Vol.2, No.4, August 2010 [14] Mahdi Khosravi, Mohammad J. Tarokh, “Dynamic Mining of Users Interest Navigation Patterns Using Naive Bayesian Method”, 978-1-42448230-6/10/$26.00 ©2010 IEEE [15] N. Sujatha, K. Iyakutty, “Refinement of Web usage Data Clustering from K-means with Genetic Algorithm”, European Journal of Scientific Research ISSN 1450-216X Vol.42 No.3 (2010), pp.464-476 [16] http://httpd.apache.org/docs/1.3/logs.html [17] http://www.w3.org/TR/WD-logfile.html [18] http://www.internetworldstats.com [19] http://www.domaintools.com/internet-statistics/

  23 International Journal of Instrumentation, Control & Automation (IJICA), Volume 1, Issue 1, 2011